How do we learn about causes of death in France?

Statistical analysis of causes of death in a particular population is based on information indicated by physicians death certificates: doctors are required to fill out this information on the certificate when notifying the civil authorities of a death. Before being used by researchers, however, the information has to be coded by INSERM’s Epidemiology Center on medical causes of death (CépiDc), which also determines the main or “underlying” cause of death, in accordance with international World Health Organization (WHO) rules.  The underlying cause of death is defined as “the disease or injury which initiated the train of morbid events leading to death.” Several other causes may be noted on the certificate: consequences or complications brought about by the underlying cause, underlying cause risk factors (for example, smoking in the case of lung cancer), or causes that may have worsened the person’s overall state of health, such as chronic diseases, obesity, or cognitive disorders. The average number of causes indicated on a death certificate is 3.5. 

No, not always. But here it is important to distinguish the doctor’s knowledge from knowledge of demographers or statisticians. According to INSERM-produced data, one in five deaths are due to an unknown or ill-defined cause. But some of those deaths are cases where the doctor knew the cause or causes of death but did not fill out the certificate correctly. It can also happen that INSERM does not receive a death certificate. This is likely to occur in cases where the death appears suspect in some way, which in turn leads to additional expert investigation in the form of autopsies or anatomical-pathological analyses that can only be conducted after official acceptance of the death certificate. The proportion of deaths due to uncertain or ill-defined causes is highest in connection with young people aged 25 to 34 (18%), ages at which what are called violent deaths (particularly suicides and accidents) occur relatively frequently. That proportion rises to about the same level for people who die at age 95 or over; here, doctors often have difficulty determining the cause of death. Two standard French expressions—“mourir de sa belle mort” (roughly, to die a natural death) and “mourir de vieillesse” (to die of old age)—illustrate this situation fairly well. 

INED has longstanding, internationally recognized expertise in causes-of-death analysis. The indicators used are fairly classic, but our analysis of long time series—extremely useful in understanding changes in mortality due to any cause and therefore of great help in forecasting life expectancy—is capable of identifying and informing on particular problems. Moreover, in the last fifteen years or so, that analysis has been enriched by multiple-causes-of-death study—that is, study that is are not limited to the underlying cause but takes account of all causes mentioned on a given death certificate. This makes particularly good sense in a context of falling mortality where people live to increasingly old ages, often reached when they have several diseases. Today we are interested in “multi-morbidity” at time of death. The increase in this type of mortality may reveal an increase in population-wide multi-morbidity, or improved survival of people with chronic diseases, or their greater vulnerability to the occurrence of certain events. This is clearly illustrated by recent history, with the 2003 heatwave and its drastic effects in France, and the COVID-19 pandemic.

In short, cause-of-death studies shed light on the overall health situation of a given country. International comparisons are often done to identify convergences and divergences. Infra-national differences and differences by sex are also investigated. And now these data can be used to produce more detailed analyses of social disparities in mortality. 

In France as elsewhere, the fall in mortality caused by cardiovascular disease was the main driver of the decline in mortality observed in the 1970s. France stands out clearly from its European neighbors for low cardiovascular mortality. It is for this reason that cardiovascular diseases are now ranked the second cause of death in France, behind cancer. Overall, cancer-caused mortality is falling in France, with the notable exception of lung cancer in women, the comparative rate of which has tripled in France since 1980. In 2022, mortality due to respiratory and nervous system diseases (a group that includes Alzheimer’s disease) returned to levels close to those found prior to the COVID-19 pandemic. The distribution of causes of death obviously differs by age group. Suicides and accidents are heavily overrepresented causes of death in young age groups. 

Sources :

Breton, D., Belliot, N., Barbieri, M., Chaput, J., & D’albis, H. (2024). Recent demographic trends in France: a singular position in the European Union, Population 79(4) [FR]

Fouillet, A., Cadillac, M., Coudin, É., & Riviera, C. (2024). Grandes causes de mortalité en France en 2022 et tendance récentes. Bulletin Epidémiologique Hebdomadaire,18 [FR]

Grippo F., Frova L., Pappagallo,M., Barbieri,M., Trias-Llimós S., Egidi V., Meslé F. & Désesquelles A. (2024). Beyond the underlying cause of death: an algorithm to study multi-morbidity at death. Population Health Metrics, 22(1), 36